Causality in a Social World: Moderation, Mediation and Spill–over

Innovative strategies for the empirical study of causal effects for social scientists

Causality in a Social World introduces a series of weighting methods, often nonparametric in nature, for addressing a wide range of causal questions while minimizing the reliance on model–based assumptions. Rather than fitting research questions to existing statistical tools regardless of their constraining assumptions, this book offers numerous application examples in which causal inquiry is driven by scientific and practical interests. 

Key features:

Demonstrates how each causal relationship can be defined in terms of potential outcomes and identified using optimal experimental designs. Compares the identification assumptions invoked by various statistical techniques for analyses of non–experimental data.  Surveys existing approaches along with presenting innovative new methodologies for causal analysis of moderation, mediation, and spill–over effects with examples from social scientific research. Provides detailed descriptions of the weighting procedures. Offers user–friendly software packages as well as syntax for the commonly used statistical programs SPSS, SAS, Stata and R. Causality in a Social World provides guidance to social scientists, graduate and post–graduate students in their empirical investigations of causality. Researchers conducting policy analysis and program evaluation in research companies or in research offices at different levels of the gov

Guanglei Hong, University of Chicago, Department of Comparative Human Development, USA